Is the idea here to capture a bad pipeline that would result in a bad setup, such as a docker container built with the wrong version?

Yes that was what I was thinking!

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 36.55%. Comparing base (73d6592) to head (abaac6b).

@@            Coverage Diff             @@
##             main     #334      +/-   ##
==========================================
+ Coverage   36.50%   36.55%   +0.04%     
==========================================
  Files          45       45              
  Lines        4240     4240              
  Branches      380      380              
==========================================
+ Hits         1548     1550       +2     
+ Misses       2670     2669       -1     
+ Partials       22       21       -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

I guess the downside to this test is that all our actions now have to run with a fetch-depth: 0 flag. I don't think that's too big of a deal, but I'm also not sure attached to this test if you think this would be a pain @castelao

I guess the downside to this test is that all our actions now have to run with a fetch-depth: 0 flag. I don't think that's too big of a deal, but I'm also not sure attached to this test if you think this would be a pain @castelao

I agree, I don't think it is a big deal to require fetch-depth: 0. Currently, the most important place to verify/check the version is when building the container, which we don't verify that today. We should attach the version of the container to the version of the library, so we can still play with the intermediate dev release but also be clear on what are we deploying in production.

I don't see a clear benefit with this test here, but if you want to add it, I support you.

Let's add it and see if it gets annoying. If it does, happy to remove it